sign in sign up
<<
Home , Enhancer (genetics), Enhancer RNA

Review Article https://doi.org/10.1038/s41594-020-0446-0

Enhancer are an important regulatory layer of the epigenome

Vittorio Sartorelli 1 and Shannon M. Lauberth 2 ✉

Noncoding RNAs (ncRNAs) direct a remarkable number of diverse functions in development and through their regula- tion of , RNA processing and . Leading the charge in the RNA revolution is a class of ncRNAs that are synthesized at active enhancers, called RNAs (eRNAs). Here, we review recent insights into the biogenesis of eRNAs and the mechanisms underlying their multifaceted functions and consider how these findings could inform future investigations into enhancer transcription and eRNA function.

he explosion of high-throughput sequencing data has Many different models for how enhancers function in con- revealed the complexity and diversity of the . trol have been proposed since their initial discovery nearly four TThese data have also unexpectedly revealed that only 1–2% decades ago19–21. Specifically, there is considerable evidence demon- of the transcriptome provides instructions for the synthesis of strating that looping of distal enhancers to their target promoters is functional , while the remaining 98–99% gives rise to a required for enhancer function (reviewed in ref. 22). For example, plethora of ncRNAs, including transfer RNAs (tRNAs), ribosomal a key study revealed that experimental induction of RNAs (rRNAs), intronic RNAs, small nuclear (sn)RNAs, small looping between the β-globin (Hbb) and its asso- nucleolar (sno)RNAs, (miRNAs) and long noncoding ciated enhancer region results in transcriptional activation of the RNAs (lncRNAs). A recent addition to the expanding list of regu- Hbb gene23. Additional analyses of the forced looping of the Hbb latory ncRNAs is the emerging class of enhancer RNAs (eRNAs), enhancer and promoter regions revealed that enhancer–promoter which are transcribed from enhancers in a tissue-specific man- contacts affect transcription by supporting an increase in the tran- ner. Increasing evidence that ncRNAs regulate scriptional burst fraction (number of transcribing ), although has fundamentally altered how the scientific community views the burst size (number of transcripts produced) was unaltered24. RNA-mediated gene regulation. These new advances in our under- The role of looping in the activation or increased rate of transcrip- standing of ncRNAs have also piqued an interest in pursuing inves- tion may be due to the binding of transcription factors, cofactors tigations of them, as the functions of the vast majority of ncRNAs and the general transcription machinery to enhancers, raising the remain to be determined. local abundance of the transcription machinery within the vicin- Enhancers are classically defined as DNA sequences that regulate ity of specific target . Enhancer looping has also been shown the gene expression networks underlying distinct cellular identi- to play a role in RNA polymerase II (RNAPII)-mediated transcrip- ties and cellular responses to environmental cues1–7. The ENCODE tional elongation. Specifically, the LIM domain–binding 1 Consortium estimates that there are >400,000 putative enhancers (LDB1), which establishes promoter–enhancer looping, plays a key encoded by the genome8. Taken together, the evidence that role in regulating the pause release of RNAPII within the β-globin enhancers account for a significant proportion of the and the gene25,26. Enhancers may also regulate target genes via transcripts identification of disease-associated genetic variants within enhancers produced from the enhancer regions themselves. eRNA produc- underscores the importance of understanding how these elements are tion is a widespread phenomenon that has been implicated in the regulated and how they function in gene expression control8,9. While regulation of gene expression in multiple types in response to sequencing technologies have advanced the ability to predict enhancer various stimuli7,27–32. The emerging prevalence of noncoding eRNAs location and activity, their functional dissection remains challenging makes them well positioned to dynamically remodel cellular tran- due, in part, to their ability to act over long and variable distances scriptomes and adds a new layer of complexity to gene regulation. with respect to their target genes and to the propensity of individual However, an issue that remains to be resolved is whether eRNAs enhancers to regulate multiple genes (reviewed in ref. 10). Additional have direct roles in gene control, as current efforts to determine challenges are that enhancer activity is dynamic and restricted to par- this face the challenge of uncoupling eRNA function from the act ticular cell types or tissues and environmental signals3–7. Moreover, the of enhancer transcription. Thus, it remains necessary to develop prediction of specific sequences that contribute to enhancer function tools to experimentally manipulate and model the direct functions is problematic because of their modest sequence conservation across of eRNAs. species11–15. It may be that a common function underlies enhancer , given that enhancer function has been shown to be con- Enhancers as functional noncoding RNA transcription units served without discernible sequence conservation16–18. Together, these Nearly a decade ago, two studies reported the intriguing finding that studies indicate the need to continue improving methodologies that enhancer regions support transcription and give rise to noncoding aid in both the functional dissection of enhancers and in uncovering eRNAs33,34. Since the discovery of eRNAs, there have been numerous the mechanisms directing their evolution. reports that eRNA is synthesized in a cell-type and signal-dependent

1Laboratory of Muscle Stem Cells & Gene Regulation, National Institute of Arthritis and Musculoskeletal and Skin (NIAMS), NIH, Bethesda, MD, USA. 2Section of Molecular , University of California, San Diego, La Jolla, CA, USA. ✉e-mail: [email protected]

Nature Structural & | www.nature.com/nsmb Review Article NAtuRe StRuctuRAL & MoLecuLAR BioLogy manner3–7. Because eRNAs are not readily detectable in steady-state transcript, S2P levels increase and S5P levels decrease69–71. On the RNA-sequencing data, their annotation depends on sequencing other hand, the elongation of enhancer transcripts is distinguished nascent RNA using approaches that include global run-on sequenc- by low levels of the S2P form of RNAPII and minimal levels of the ing (GRO-seq)3,5–7,27,35–40, precision run-on nuclear sequencing elongation-specific mark H3K36me3, both of which are (PRO-seq)41 and cap analysis gene expression (CAGE)42,43. These enriched within gene bodies of lncRNAs and mRNAs55,72. Low lev- nascent transcription assays have been instrumental in uncovering els of S2P are consistent with the enrichment of the Tyr1P form of a wide array of ncRNAs, including long noncoding RNAs, enhancer RNAPII that is specifically found at active enhancers and not at RNAs, promoter upstream transcripts and upstream antisense the sense strand of gene promoters73,74. The prevalence of Tyr1P RNAs. Indeed, global annotation analyses have revealed that eRNA coincides with the production of eRNAs and promoter-directed transcripts account for a large proportion of initiation events in the upstream antisense transcripts (PROMPTs), both of which are rela- transcriptome, with approximately 40,000–65,000 eRNAs expressed tively unstable due to exosome-mediated degradation73. Notably, in human cells4,44. The annotation of such transcripts in Drosophila Tyr1P is required for transcription termination control in yeast68,75 melanogaster45,46 and Caenorhabditis elegans47 reinforces the finding and, more recently, has been shown to prevent transcription that eRNAs are a common feature of active enhancers in metazoans. readthrough at mammalian gene ends76. However, no apparent con- eRNAs are produced from active enhancers that share several tribution of Tyr1P to the regulation of transcription readthrough is features: (i) an open chromatin state, reflected by the presence of observed at enhancers76. Thus, Tyr1P accumulation may contribute DNase hypersensitive sites (DHSs); (ii) binding of transcription to different functional consequences at promoters and enhancers. factors and cofactors, including the histone acetyltransferase p300 Further demonstrating a role for elongation in enhancer regula- and cAMP –binding protein (CBP); and (iii) the tion, the RNAPII-associated transcription elongation factor SPT6 co-occurrence of histone H3 lysine 4 monomethylation () contributes to the recruitment of Integrator to lncRNA genes. The and histone H3 lysine 27 acetylation ()48–52. While these Integrator subunit INTS3 is enriched at enhancer regions that gener- features have guided enhancer identification, not all active enhanc- ate bidirectional eRNAs, and SPT6 depletion substantially decreases ers that share these features direct eRNA production53. Also, not INTS3 recruitment and eRNA production77,78. The elongation factor all active enhancers support comparable levels of transcriptional ELL3 has also been shown to regulate occupation of enhancers by activity. Higher levels of eRNA synthesis have been shown to cor- RNAPII79. Additional studies are needed to determine whether dis- relate with an increased ratio of histone H3 lysine 4 trimethyl- tinct mechanisms regulate mRNA and eRNA elongation and to fully ation () to H3K4me1 and increased levels of RNAPII understand the factors underlying the regulation of transcription occupancy37,54,55. These findings suggest that measuring H3K4me1 and termination at enhancers. accumulation alone may underestimate the number of functional enhancers and that H3K4me3 may be a superior predictor of the Classification of noncoding enhancer RNAs level of enhancer activity. Thus, identifying additional features that A comprehensive definition of eRNAs has not yet been formulated, can be used to predict enhancers will improve the identification as annotated eRNAs are defined as both (i) short, bidirectional, of functionally active enhancers and bona fide eRNAs. In turn, non-polyadenylated, non-spliced and unstable and (ii) unidirec- eRNAs themselves may prove to be the most reliable predictor of tionally transcribed, spliced, polyadenylated and stable33,80,81 (Fig. 1). enhancer activity. While the majority of eRNAs are not polyadenylated or spliced, they tend to be capped, as shown by nuclear run-on assays followed Regulation of enhancer transcription by 5′ cap sequencing (5′GRO-seq) and CAGE4,7,37. Consistent with Enhancers and promoter regions of protein-encoding genes share their low abundance and instability, eRNAs are predominantly similar properties and rules for transcription initiation. Core pro- localized in the nucleus and chromatin-bound fractions4,33,34,36,82,83. moter sequences, such as the TATA box, spacing and The early transcription termination of eRNAs is regulated by the the assembly of general transcription factors (TFIID/RNAPII) Integrator complex in a manner that is probably dependent on and cofactors (, p300) are observed at both enhancers and termination-triggering (pA)-like signals6,37,84. In promoters55,56. Consistent with this, several studies have revealed addition, the high turnover rate of eRNAs is mediated by the nuclear that, similar to promoter regions, transcription at enhancers pre- RNA exosome complex4,84–86. The length of eRNA transcripts is pre- dominantly occurs in a bidirectional manner4,37,46,55,57–59. Enhancers dicted to be less than 150 , based on the identification of and promoters have also been shown to be functionally inter- transcription termination sites within ~150 nucleotides of the tran- changeable in supporting RNAPII initiation: intragenic enhancers scription start site of enhancer loci46,87. Notably, current predictions behave as alternative promoters and can functionally substitute for of eRNA length that stem from PRO-seq and Start-seq datasets do promoters to drive mRNA and long noncoding RNA transcrip- not exclude the possibility of longer eRNA transcripts46,87. tion60. Studies in Drosophila have also revealed that bidirectionally The current predictions regarding eRNA length have led to transcribed enhancers behave as weak promoters and, conversely, their classification as a distinct family of lncRNAs. Nonetheless, that bidirectionally transcribed promoters can function as strong lncRNAs are distinguished from eRNAs not only by their length, enhancers53. These findings blur the classical definitions of pro- which is >200 nucleotides, but also by their processing, which is moters and enhancers and raise the possibility that noncoding consistent with their stability88–90. Importantly, however, the defini- transcripts generated at both regulatory regions may be functional tions of lncRNAs and eRNAs are not mutually exclusive, and it is (reviewed in refs. 61,62). thus likely that some annotated eRNAs are actually lncRNAs and By comparison, there are discernible differences between the vice versa. This inherent complexity in classifying eRNAs is due, transcriptional elongation phases of eRNA and mRNA transcript in part, to the difficulty in uncoupling specific lncRNAs from their production. These differences primarily reflect the well-defined associated enhancer elements33,81. For example, a recent study found cycles of RNAPII activity, which correlate with phosphorylation that genomic regions defined by bidirectional transcription and and dephosphorylation of the carboxyl terminal domain (CTD) of enhancer features (termed eRNA-producing centers, or EPCs) are mammalian RNAPII63–65. The CTD domain consists of a heptam- located in proximity to the transcription start sites of lncRNAs80. eric sequence (YSPTSPS) repeated 52 times that is phosphorylated The ability of the EPCs to drive lncRNA production was shown to at S2, S5, S7, T4 and Y166–68. The general model of CTD phosphory- be correlated with higher enhancer activity and is in part linked to lation during transcription is that the CTD becomes enriched with lncRNA maturation and the presence of evolutionarily conserved S5P at the 5′ ends of coding regions, and, as RNAPII elongates a U1 splicing motifs80. As is the case for bidirectional promoters91, the

Nature Structural & Molecular Biology | www.nature.com/nsmb NAtuRe StRuctuRAL & MoLecuLAR BioLogy Review Article a 2D-eRNA b 1D-eRNA revealed that inhibition of transcriptional elongation at enhancers affects H3K4me1 and H3K4me2 deposition and is independent of Short bidirectional Long unidirectional eRNAs83. Also, transcription at intragenic enhancers, and not the eRNA itself, was found to interfere with and attenuate host gene expression during embryonic stem cell differentiation97. Similarly, a prior report revealed a requirement for the lncRNA Lockd in regu- Unspliced Spliced lating gene expression in a manner that is dependent on its associ- ated enhancer elements but not the Lockd transcripts92. Moreover, not all active enhancers support eRNA production, suggesting that eRNAs are not functionally important at all enhancers53. However, this remains a matter of continued debate, as it is presently unclear

Non-polyadenylated Polyadenylated whether these enhancers produce low levels of eRNAs that are not readily detectable with current methods53. Nevertheless, a growing number of studies have shown that a subset of eRNAs are required AAAA to support the expression of cognate target genes, and we discuss 7,27–32 Cis Trans these various eRNA functions in detail below. A variety of experimental methods continue to advance our understanding of eRNA function. Perturbing nuclear decay path- ways by knocking down the core components Exosc3 and nuclear RNase Exosc10 of the RNA exosome results in stabilization of eRNAs and permits their functional assessment4,85. It is impor- tant to note that, upon RNA exosome ablation, eRNA-expressing regions show an accumulation of R-loops, which is consistent with Fig. 1 | Molecular features that define enhancer RNAs. Schematic diagram 56 the involvement of the RNA exosome in resolving R-loops at active to depict the differences between annotated eRNAs . Distinct transcripts enhancers85. Another methodology to interrogate the functions of synthesized at enhancers are frequently classified as bidirectional eRNAs includes CRISPR-Display, in which eRNAs are tethered to (2D-eRNAs) or unidirectional (1D-eRNAs). a, The majority of enhancer catalytically dead Cas9 (dCas9) for targeting to specific genomic transcripts are 2D-eRNAs that are comparatively short, non-polyadenylated loci98. Also, loss-of-function studies, such as eRNA knockdown and non-spliced and function in cis4,33,34,36,82,83. b, 1D-eRNAs are longer, 33,80,81 29 analyses, could be used to target the enhancer region (deletion or unidirectional, polyadenylated and spliced and can function in trans . insertion of genetic elements) or the enhancer-directed transcript (RNAi-mediated knockdown or incorporation of antisense oligo- nucleotides (ASO)). However, two recent studies have revealed that enrichment of U1 splicing sites at EPCs may prevent premature ter- ASO-mediated knockdown is limited by its inability to discriminate mination and permit productive elongation by RNAPII to generate between RNA function and the act of transcription because incorpo- lncRNAs. It will therefore be of interest to determine whether this ration of ASOs results in premature transcription termination99,100. same mechanism also applies at enhancers and whether lncRNAs Several studies have also explored eRNA functions by employing and eRNAs establish a positive transcriptional feedback loop that inhibitors of transcription elongation, such as actinomycin D and regulates eRNA and lncRNA synthesis. It should be noted, however, flavopiridol28,36,83,101. These methods used to probe the functional that only a minority (3–5%) of the total EPCs are in proximity to consequences of eRNAs are often paired with approaches that can lncRNAs80. Similarly, the annotated lncRNAs linc-p21 (upstream of assess where in the genome the eRNAs are functioning. Specifically, Cdkn1a) and Lockd (downstream of Cdkn1b) regulate gene expres- chromatin isolation by RNA purification (ChIRP)-seq is a power- sion in a manner that is dependent on enhancer elements that are ful method to determine the genomic sites bound by eRNAs27,102. mapped to lncRNA loci rather than to the lncRNAs themselves92,93. In addition, single-molecule fluorescence in situ hybridization In addition, reports that enhancer-associated lncRNAs function in a (smFISH), which provides a quantitative assessment of RNAs that manner similar to eRNAs and, conversely, that eRNAs derived from are localized at distinct transcriptionally active regions, can be super enhancers function as lncRNAs, reinforce the idea that these employed to investigate the cellular localization of eRNAs29,103. two classes of ncRNAs are not mutually exclusive94,95. These studies While evidence for eRNA function is rapidly emerging, it is yet are also consistent with the notion that lncRNAs may have evolved to be determined whether eRNAs act in cis versus in trans. Given from eRNAs. This suggestion is based on the presence of splicing that the majority of eRNAs are relatively unstable, it is not sur- and 3′ processing sequences in the vicinity of specific enhancers, prising that they are expected to act in cis4,33,34,36,82,83. Several recent which could promote eRNA stabilization and the evolutionary selec- studies have shown that eRNAs function in cis by demonstrating tion of new trans-acting functions96. Further analyses are needed eRNA-dependent transcriptional regulation of mRNAs produced to unravel the respective functions associated with particular DNA from loci adjacent to the corresponding eRNA-producing enhancer regions and the RNA transcripts they . Moreover, updates to regions7,27–29,85. Specific eRNAs have been shown to function by inter- our systems are needed to distinguish between eRNAs and lncRNAs acting with CBP and BRD4 at the enhancers where these eRNAs are and to resolve current investigations into the functional specializa- produced and transcriptional regulators are localized28,104. However, tion of these classes of ncRNAs. eRNAs have also been found to relocate to chromosomal regions distinct from those they are produced from to perform functions Functional roles of enhancer transcription and eRNAs in trans29. A distal regulatory MyoD enhancer (distal regulatory While eRNAs have become a hallmark of active enhancers, it region, or DRR) is transcribed into the DRR eRNA that medi- remains to be resolved whether enhancer transcription, eRNAs ates cohesin recruitment and promotes Myogenin gene expression themselves, or both, are important for enhancer activity. Studies in trans to control myogenic differentiation29. Similarly, an eRNA have demonstrated that enhancer transcription is important for synthesized adjacent to the kallikrein related peptidase 3 (KLK3) maintaining an open chromatin state that is readily accessible to gene functions in trans to enhance androgen receptor–dependent transcription factors and cofactors83. Specifically, investigation of gene expression in human prostate cancer105. In both cases, the DRR the formation of de novo enhancers during macrophage activation and KLK3 eRNAs are polyadenylated, which raises the possibility

Nature Structural & Molecular Biology | www.nature.com/nsmb Review Article NAtuRe StRuctuRAL & MoLecuLAR BioLogy that post-transcriptional regulation of eRNAs contributes to both a Stabilization of enhancer–promoter looping and establishment increased eRNA stability and eRNA function in trans. Alternatively, of chromatin accessibility the aforementioned eRNAs may be lncRNAs operating in trans. Cis Trans Future studies focused on establishing direct links between eRNAs 1 and gene loci will aid in determining features that dictate whether Chromosome 1 eRNAs function in cis or in trans. Promoter eRNA Promoter Cohesin eRNA eRNAs contribute to gene control by altering the chromatin Cohesin Enhancer environment. Several functional studies have revealed that eRNAs Enhancer play an important role in regulating gene expression by modulat- Chromosome 2 ing chromatin structure and function. Specifically, eRNAs direct chromatin accessibility at protein-coding promoters106. In addi- tion, eRNAs function in cis to contribute to the dynamic stabiliza- b Regulation of chromatin landscape tion of enhancer–promoter looping27,36,105 and in trans to regulate chromatin-remodeling events that modulate CBP complex assembly and gene regulation during myogenic differen- tiation29 (Fig. 2a). eRNAs also regulate the chromatin landscape through their interactions with epigenetic modifying that Acetylation deposit (‘write’) post-translational modifications on the histone eRNA tails. Specifically, eRNAs have been shown to promote gene expres- sion through their ability to augment histone acetylation at enhanc- RNAPII 104 ers through interactions with the histone acetyltransferase CBP Enhancer (Fig. 2b). Future analyses are required to extend the understanding of eRNAs in the regulation of chromatin structure and function. Fig. 2 | eRNA regulation of enhancer–promoter interactions and the epigenetic state of chromatin. a, Schematic diagram depicting the roles eRNAs interact with transcriptional regulators to control gene of eRNAs acting in cis to regulate specific enhancer–promoter looping expression. To advance our understanding of how eRNAs control by supporting cohesin binding at enhancers27,106 (left) and influencing gene regulation, numerous studies have focused on the identifi- chromatin accessibility by modulating cohesin complex recruitment cation and classification of the eRNA-interacting proteome. The in trans to regulate gene expression (right). b, Schematic showing that methods most frequently employed to identify direct binding part- eRNAs stimulate the catalytic activity of the histone acetyltransferase ners and physiological associations of eRNAs include RNA elec- CBP to regulate increased deposition of histone acetylation, which in turn trophoretic mobility shift assays and RNA immunoprecipitation contributes to greater enhancer activation104. (UV-RIP), respectively. Purification of target eRNAs complexed with proteins is often performed by using biotinylated antisense to capture the RNAs, coupled with mass spectrom- many types of human . eRNAs are expressed across human etry (RAP-MS). These assays have identified eRNAs that regulate tissues3,5,21,27,31, supporting their potential relevance as thera- gene expression via interactions with RNAPII, various transcription peutic targets and biomarkers. The biggest barrier to moving these factors and cofactors. eRNAs have been shown to interact with Yin molecules to the clinic as a new generation of actionable targets is Yang 1 (YY1) to increase recruitment of this transcriptional regula- the lack of knowledge of their biological functions. However, recent tor at enhancers107 (Fig. 3a, left). Similarly, eRNAs increase RNAPII studies have begun to uncover roles for eRNAs in the regulation occupancy at protein-coding loci28,106. It remains to be determined of tumor-promoting gene expression programs. The ARIEL eRNA whether transcription factor and cofactor trapping at enhancers has recently been shown to promote activation of an oncogenic and gene loci is a widespread mechanism underlying eRNA func- gene-expression program in T-cell acute lymphoblastic leukemia109. tion. In addition to regulating RNAPII binding, eRNAs also regu- eRNAs have also been shown to regulate the chromatin interactions late RNA Pol II pause release by acting as a decoy for the negative and transcriptional activities of BRD4, a potent cancer dissemina- elongation factor (NELF) complex30 and through activation of the tor, that are required to regulate a subset of tumor-promoting genes positive transcription elongation factor b (P-TEFb) complex108 in response to chronic TNF-α signaling in colon cancer28 (Fig. 3). (Fig. 3b). Several recent studies have also linked eRNA functions Beyond their roles in the regulation of gene expression in can- to their interactions with several general cofactors, including cohe- cer, eRNAs are also linked to the maintenance of genome stability. sin27,36, Mediator31, CBP104 and BRD4 (ref. 28). eRNA interactions Increased levels of eRNAs in human cancers have been shown to with cohesin and CBP augment enhancer activation through the promote the formation of three-stranded DNA:RNA regulation of chromatin looping and increased deposition of his- hybrids (R-loops) that interfere with DNA replication and, in tone acetylation, respectively27,36,104. eRNAs directly interact with turn, induce chromosome rearrangements and genome instabil- BRD4 through its bromodomains (BDs) to promote greater bind- ity (reviewed in ref. 110). Consistent with the possibility that aber- ing of BRD4 to acetylated , which, in turn, contributes to rant eRNA synthesis contributes to tumorigenesis is the finding the maintaining enhancers in an active state28 (Fig. 3a, right). Lastly, that enhancer transcription leads to activation-induced cytidine eRNA interactions with Mediator are required to support transcrip- deaminase (AID) mistargeting, which promotes genome insta- tional activation by promoting chromatin looping31. Importantly, bility and malignancy111. Similarly, , including mutant all of these studies are consistent with eRNAs exhibiting their , have been shown to drive potent levels of eRNA synthesis in functional roles in collaboration with different binding partners human colon cancer cell lines3, which may lead to aberrant R-loop involved in transcriptional regulation. formation and the loss of genome stability. Future studies that examine the mechanisms underlying the accumulation of eRNAs, eRNAs in tumor-promoting gene regulation and genomic insta- R-loop formation and gene expression changes, and the con- bility in cancer. Since enhancers are known to control the selec- nections between these events in normal versus cancer cells, will tion and maintenance of hundreds of cell types, it is not surprising help to address remaining questions in the enhancer biology field. that enhancer misregulation has emerged as a driving force behind One unresolved issue relates to how enhancer regions are able to

Nature Structural & Molecular Biology | www.nature.com/nsmb NAtuRe StRuctuRAL & MoLecuLAR BioLogy Review Article a TF and chromatin reader trapping were examined for direct interactions with CBP104 and BRD428, sug- Acetylation gesting that additional analyses are needed to investigate the speci- ficity of interactions between eRNAs and their respective binding partners. Consistent with the possibility that eRNAs may function

YY1 through specific sequence motifs is the finding that a similar motif is present in a subset of androgen-receptor-regulated eRNAs that Enhancer promotes transcription by activating P-TEFb108. In addition, cur- rent analyses cannot rule out the contributions of eRNA length, structure, and conformational flexibility. A further possibility that remains to be explored is whether post-translational modifica- BRD 4 tion of eRNAs contributes to the regulation and function of eRNA binding partners. To this end, deposition of 5-methylcytosine on eRNA eRNAs supports the functions of PGC1-α during meta- 112 RNA Pol II bolic stress . Chemical modification of eRNAs may thus provide an additional layer of regulation in the interactions between eRNAs Enhancer and binding partners and may thereby have important implications for enhancer and gene control. Interestingly, eRNAs were shown to interact with binding partners through noncanonical RNA binding regions (RBRs), which are dis- b Transition of RNAPII into productive elongation tinct from the well-characterized RNA-interacting domains, includ- 113 eRNA ing the RNA-recognition motif (RRM) , the hnRNPK-homology domain (KH)114 and the double-stranded RNA binding domain NELF (dsRBD)115. For example, eRNAs have recently been shown to Acetylation interact with the highly conserved acetyl-lysine-binding BDs of the

P-TEFb BET family members and the single BDs of the non-BET proteins 28 P BRG1 and BRD7 . This study raises the interesting questions of RNA Pol II how prevalent eRNA interactions with chromatin reader domains Promoter are and what the significance of such interactions is in the regula- tion of chromatin and in gene control. Several RBRs in CBP/p300 were predicted to interact with eRNAs104; however, this was attrib- Fig. 3 | eRNAs modulate the chromatin interactions of transcriptional uted to a single CBP RBR that resides within the catalytic histone regulators. Schematic illustrating the role of eRNAs in directly regulating acetyltransferase domain. eRNA interactions with this domain were transcription. a, eRNAs regulate the enhancer occupancy of the shown to contribute to the displacement of the CBP activation loop transcription factor YY1107 and the transcriptional coactivator BRD4 (ref. 28). that blocks substrate binding to the active site to allow for enhanced b, eRNAs promote RNAPII chromatin engagement and the transition from histone acetyltransferase activity104. Moreover, further analyses will RNAPII pause release to productive elongation by acting as a decoy for be needed to investigate the significance of eRNA interactions with NELF30 and by contributing to the activation of P-TEFb108. (i) both canonical and noncanonical RNA binding regions; (ii) mul- tiple domains within a single protein, as was predicted for CBP104; support high levels of eRNA synthesis while remaining uninhibited and (iii) multiple subunits within a multiprotein complex. Taken by increased formation of R-loops. Another possibility that remains together, these studies allude to the exciting possibilities for advanc- to be explored is whether single polymorphisms (SNPs) ing our mechanistic understanding of the biological and biochemi- that are localized within enhancers alter or disrupt eRNA functions; cal roles of eRNAs and their corresponding binding partners. Yet a recent high-throughput analysis that identified SNPs that alter additional studies are required to obtain a comprehensive list of the regulatory functions of enhancers and promoters is consistent eRNA binding partners and to uncover the molecular mechanisms with this possibility9. Genes that exhibit transcriptional burst fre- underlying these interactions. Such analyses are likely to have broad quency differences were found to have higher densities of SNPs in implications, based on widespread eRNA synthesis in different cell their enhancer but not promoter regions. Identification of enhancer types and because a large-scale analysis of eRNA binding partners transcription ‘signatures’ and elucidation of the mechanisms has not yet been performed. underlying the precise functions of eRNAs in cancers will pave the way for the potential use of eRNAs as diagnostic markers and Future perspectives therapeutic targets. The widespread involvement of ncRNAs in the regulation of gene expression suggests that a great deal remains to be discovered about Mechanisms underlying the functions of eRNAs the functional significance of these important molecules. On the Recent studies have demonstrated that eRNA functions are basis of current studies, the causal roles of eRNAs are likely to be largely mediated through interactions with eRNA binding part- important for the regulation of a subset of enhancers and target loci. ners27,28,31,36,104,105. Several studies suggest that RNA sequence is prob- Nonetheless, additional evidence is required to definitively iden- ably not involved in contributing to the specificity of eRNA–protein tify eRNAs and to elucidate their biological roles. Notably, these interactions. For example, both CBP104 and BRD428 were found to advances will require improvements to methodologies that cannot bind a broad spectrum of RNAs, which is consistent with these fac- currently distinguish between the roles of enhancer transcription tors binding RNA in a non-sequence-specific manner. However, and eRNAs. As such, approaches that evaluate enhancer-mediated both of these studies suggest that the specificity of these eRNA transcription dynamics and the direct roles of eRNAs are needed interactions stems from eRNAs interacting with CBP and BRD4 in to advance our understanding of enhancer regulation and function. an enhancer-specific manner28,104. This property may be exploited This includes cell-free systems, which could be used to manipulate by the pervasive binding of transcriptional cofactors throughout the and model the roles of eRNAs and eRNA-associating binding pro- genome, such that RNAs could stimulate activity locally, indepen- teins and thus close an important gap in our knowledge of enhancer dently of RNA sequence. Nonetheless, only a few specific eRNAs and gene regulation.

Nature Structural & Molecular Biology | www.nature.com/nsmb Review Article NAtuRe StRuctuRAL & MoLecuLAR BioLogy

Biomolecular RNP condensate eRNA binding partners is tailored to specific enhancers to con- trol distinct gene expression programs and cellular processes. It will also be important to consider the recently identified role of eRNA-containing protein complexes in the regulation and func- Chromatin tion of the RNA-dependent architecture that directs the formation remodeling 116 factors and integrity of phase-separated condensates (Fig. 4). Several questions that remain to be addressed include the following: how do eRNAs contribute to changes in the physical properties of conden- Coactivator Acetylation sates, and can eRNAs specifically regulate the sorting of proteins and regulatory DNA elements into specific condensates? Which eRNA RNA Pol II motifs are sufficient for the formation of phase separated particles? TF Finally, it remains to be determined whether eRNA sequences that show increased affinity for binding partners exist. High-affinity Enhancer binding sites and the potential for cooperative spreading of bind- eRNA ing partners on eRNAs, which has been shown for mRNAs, could play an important role in facilitating multivalent interactions, which contribute to phase separation117. In addition, eRNAs provide the potential for multivalency in RNA-mediated chromatin associa- tions28, as demonstrated by the recent finding that eRNAs interact Biomolecular condensate with BDs to facilitate enhanced chromatin associations and the downstream transcription-associated functions of BRD4. Histone reader proteins tend to form weak interactions with chromatin Fig. 4 | The role of eRNAs in regulating condensate assembly on through their recognition of and direct interactions with particu- enhancers. Schematic representation of an enhancer-RNA-dependent lar histone marks. This model would be consistent with previous ribonucleoprotein (eRNP) complex exhibiting properties of a findings that the overall affinity and specificity of chromatin bind- phase-separated condensate. A growing number of studies suggest ing is enhanced when multiple chromatin-interacting domains in a that the formation of liquid-like droplets or phase condensates may single protein or within a multiprotein complex are able to engage be a general mechanism to compartmentalize biochemical reactions with chromatin118. that support cellular activities (reviewed in refs. 119,120). In support of the In summary, an explosion of studies in the last decade has identi- phase-separation model, recent studies have revealed that the recruitment fied noncoding enhancer transcripts that are produced from a wide of multimolecular assemblies to super-enhancers (SEs) imparts the array of active enhancers in multiple cell types and tissues. Building ability to form phase condensates to SEs119. It has also been demonstrated on this foundation, the next decade will be an exciting time for that phase separation at SEs plays a key role in transcriptional control uncovering the mechanisms that control enhancer transcription through the compartmentalization of transcriptional regulators119,121–123. and eRNA functions in gene regulation and biological processes. Emerging evidence also shows that eRNAs produced from ligand-activated Progress toward this goal will be the next fundamental step in enhancers regulate enhancer activation by controlling the diffusion advancing our understanding of gene regulation. properties of phase-separated components116. Specifically, eRNAs and eRNA binding partners with intrinsically disordered regions were found to Received: 25 July 2019; Accepted: 7 May 2020; form ribonucleoprotein (eRNP) complexes at enhancers, which exhibited Published: xx xx xxxx properties of dynamic phase-separated condensates116. This same study revealed that eRNAs play a pivotal role in regulating the fluid properties References of the condensates by promoting a more dynamic and liquid-like state 1. Plank, J. L. & Dean, A. Enhancer function: mechanistic and genome-wide rather than a gel-like state, which is required for transcriptional activity116. insights come together. Mol. Cell 55, 5–14 (2014). Conversely, phase separation also regulates enhancer activity and eRNA 2. Long, H. K., Prescott, S. L. & Wysocka, J. Ever-changing landscapes: production. Phase separation has recently been shown to regulate transcriptional enhancers in development and evolution. Cell 167, 1170–1187 (2016). the spatial interactions between enhancers that are required for both 3. Rahnamoun, H. et al. Mutant p53 shapes the enhancer landscape of cooperative ligand-mediated enhancer activation and the production of cancer cells in response to chronic immune signaling. Nat. Commun. 8, eRNAs116. Together, these recent findings reveal the importance of further 754 (2017). dissecting the relationship between eRNAs and phase separation to 4. Andersson, R. et al. An atlas of active enhancers across human cell types and tissues. Nature 507, 455–461 (2014). understand key elements of their functional mechanisms. 5. Franco, H. L., Nagari, A. & Kraus, W. L. TNFα signaling exposes latent estrogen receptor binding sites to alter the breast cancer cell transcriptome. Mol. Cell 58, 21–34 (2015). Future directions will also involve cataloging additional eRNAs, 6. Lai, F., Gardini, A., Zhang, A. & Shiekhattar, R. Integrator mediates the biogenesis of enhancer RNAs. Nature 525, 399–403 (2015). determining their cellular localization and expression patterns and 7. Lam, M. T. et al. Rev-Erbs repress macrophage gene expression by exploring their mechanisms of action. Toward this goal, it will be inhibiting enhancer-directed transcription. Nature 498, 511–515 (2013). necessary to define a more comprehensive list of eRNA binding 8. ENCODE Project Consortium. An integrated encyclopedia of DNA partners and to determine the regulatory parameters underlying elements in the . Nature 489, 57–74 (2012). their interactions with eRNAs. Categorical data analyses and sort- 9. van Arensbergen, J. et al. High-throughput identifcation of human SNPs afecting regulatory element activity. Nat. Genet. 51, 1160–1169 (2019). ing of eRNA binding partners into functional and structural classes 10. Schoenfelder, S. & Fraser, P. Long-range enhancer-promoter contacts in will provide new insights into how eRNAs interact with eRNA bind- gene expression control. Nat. Rev. Genet. 20, 437–455 (2019). ing partners. These analyses could also provide new insights into 11. Villar, D. et al. Enhancer evolution across 20 mammalian species. Cell 160, whether eRNA binding partners display specificity for different 554–566 (2015). RNA sequences and structural features and, if so, whether either 12. Arnold, C. D. et al. Quantitative genome-wide enhancer activity maps for fve Drosophila species show functional enhancer conservation and turnover or both of these possibilities impart the eRNA–protein complexes during cis-regulatory evolution. Nat. Genet. 46, 685–692 (2014). with varying functions. It will also be important to address the 13. Schmidt, D. et al. Five-vertebrate ChIP-seq reveals the evolutionary question of whether the activity of eRNAs and their corresponding dynamics of transcription factor binding. Science 328, 1036–1040 (2010).

Nature Structural & Molecular Biology | www.nature.com/nsmb NAtuRe StRuctuRAL & MoLecuLAR BioLogy Review Article

14. Blow, M. J. et al. ChIP-seq identifcation of weakly conserved heart 45. Meers, M. P. et al. Transcription start site profling uncovers divergent enhancers. Nat. Genet. 42, 806–810 (2010). transcription and enhancer-associated RNAs in . 15. May, D. et al. Large-scale discovery of enhancers from human heart tissue. BMC Genomics 19, 157 (2018). Nat. Genet. 44, 89–93 (2011). 46. Henriques, T. et al. Widespread transcriptional pausing and elongation 16. Yang, S. et al. Functionally conserved enhancers with divergent sequences control at enhancers. Genes Dev. 32, 26–41 (2018). in distant vertebrates. BMC Genomics 16, 882 (2015). 47. Chen, R. A.-J. et al. Te landscape of RNA polymerase II transcription 17. Chatterjee, S., Bourque, G. & Lufin, T. Conserved and non-conserved initiation in C. elegans reveals promoter and enhancer architectures. enhancers direct tissue specifc transcription in ancient specifc Genome Res. 23, 1339–1347 (2013). developmental control genes. BMC Dev. Biol. 11, 63 (2011). 48. Chepelev, I., Wei, G., Wangsa, D., Tang, Q. & Zhao, K. Characterization of 18. Hare, E. E., Peterson, B. K., Iyer, V. N., Meier, R. & Eisen, M. B. Sepsid genome-wide enhancer-promoter interactions reveals co-expression of even-skipped enhancers are functionally conserved in Drosophila despite interacting genes and modes of higher order chromatin organization. lack of sequence conservation. PLoS Genet. 4, e1000106 (2008). Cell Res. 22, 490–503 (2012). 19. Beagrie, R. A. & Pombo, A. Gene activation by metazoan enhancers: 49. Heintzman, N. D. et al. Histone modifcations at human enhancers refect diverse mechanisms stimulate distinct steps of transcription. Bioessays 38, global cell-type-specifc gene expression. Nature 459, 108–112 (2009). 881–893 (2016). 50. Creyghton, M. P. et al. Histone H3K27ac separates active from poised 20. Furlong, E. E. M. & Levine, M. Developmental enhancers and chromosome enhancers and predicts developmental state. Proc. Natl Acad. Sci. USA 107, topology. Science 361, 1341–1345 (2018). 21931–21936 (2010). 21. Catarino, R. R. & Stark, A. Assessing sufciency and necessity of enhancer 51. Rada-Iglesias, A. et al. A unique chromatin signature uncovers early activities for gene expression and the mechanisms of transcription developmental enhancers in . Nature 470, 279–283 (2011). activation. Genes Dev. 32, 202–223 (2018). 52. Wang, D. et al. Reprogramming transcription by distinct classes of 22. Krivega, I. & Dean, A. Enhancer and promoter interactions—long distance enhancers functionally defned by eRNA. Nature 474, 390–394 (2011). calls. Curr. Opin. Genet. Dev. 22, 79–85 (2012). 53. Mikhaylichenko, O. et al. Te degree of enhancer or is 23. Deng, W. et al. Controlling long-range genomic interactions at a native refected by the levels and directionality of eRNA transcription. Genes Dev. locus by targeted tethering of a looping factor. Cell 149, 1233–1244 (2012). 32, 42–57 (2018). 24. Bartman, C. R., Hsu, S. C., Hsiung, C. C.-S., Raj, A. & Blobel, G. A. 54. Pekowska, A. et al. H3K4 tri-methylation provides an epigenetic signature Enhancer regulation of transcriptional bursting parameters revealed by of active enhancers. EMBO J. 30, 4198–4210 (2011). forced chromatin looping. Mol. Cell 62, 237–247 (2016). 55. Koch, F. et al. Transcription initiation platforms and GTF recruitment at 25. Sawado, T., Halow, J., Bender, M. A. & Groudine, M. Te β-globin locus tissue-specifc enhancers and promoters. Nat. Struct. Mol. Biol. 18, control region (LCR) functions primarily by enhancing the transition 956–963 (2011). from transcription initiation to elongation. Genes Dev. 17, 56. Natoli, G. & Andrau, J. C. Noncoding transcription at enhancers: general 1009–1018 (2003). principles and functional models. Annu. Rev. Genet. 46, 1–19 (2012). 26. Song, S.-H. et al. Multiple functions of Ldb1 required for β-globin 57. Flynn, R. A., Almada, A. E., Zamudio, J. R. & Sharp, P. A. Antisense RNA activation during erythroid diferentiation. Blood 116, 2356–2364 (2010). polymerase II divergent transcripts are P-TEFb dependent and substrates 27. Li, W. et al. Functional roles of enhancer RNAs for oestrogen-dependent for the RNA exosome. Proc. Natl Acad. Sci. USA 108, 10460–10465 (2011). transcriptional activation. Nature 498, 516–520 (2013). 58. Scruggs, B. S. et al. Bidirectional transcription arises from two distinct hubs 28. Rahnamoun, H. et al. RNAs interact with BRD4 to promote enhanced of transcription factor binding and active chromatin. Mol. Cell 58, chromatin engagement and transcription activation. Nat. Struct. Mol. Biol. 1101–1112 (2015). 25, 687–697 (2018). 59. Preker, P. et al. PROMoter uPstream Transcripts share characteristics with 29. Tsai, P.-F. et al. A muscle-specifc enhancer RNA mediates cohesin mRNAs and are produced upstream of all three major types of mammalian recruitment and regulates transcription in trans. Mol. Cell 71, promoters. Nucleic Acids Res. 39, 7179–7193 (2011). 129–141.e8 (2018). 60. Kowalczyk, M. S. et al. Intragenic enhancers act as alternative promoters. 30. Schaukowitch, K. et al. Enhancer RNA facilitates NELF release from Mol. Cell 45, 447–458 (2012). immediate early genes. Mol. Cell 56, 29–42 (2014). 61. Andersson, R. & Sandelin, A. Determinants of enhancer and promoter 31. Lai, F. et al. Activating RNAs associate with Mediator to enhance chromatin activities of regulatory elements. Nat. Rev. Genet. 21, 71–87 (2020). architecture and transcription. Nature 494, 497–501 (2013). 62. Kim, T. K. & Shiekhattar, R. Architectural and functional commonalities 32. Ivaldi, M. S. et al. Fetal γ-globin genes are regulated by the BGLT3 long between enhancers and promoters. Cell 162, 948–959 (2015). noncoding RNA locus. Blood 132, 1963–1973 (2018). 63. Hirose, Y. & Manley, J. L. RNA polymerase II and the integration of nuclear 33. De Santa, F. et al. A large fraction of extragenic RNA Pol II transcription events. Genes Dev. 14, 1415–1429 (2000). sites overlap enhancers. PLoS Biol. 8, e1000384 (2010). 64. Proudfoot, N. J., Furger, A. & Dye, M. J. Integrating mRNA processing with 34. Kim, T.-K. et al. Widespread transcription at neuronal activity-regulated transcription. Cell 108, 501–512 (2002). enhancers. Nature 465, 182–187 (2010). 65. Shatkin, A. J. & Manley, J. L. Te ends of the afair: capping and 35. Franco, H. L. et al. Enhancer transcription reveals subtype-specifc gene polyadenylation. Nat. Struct. Biol. 7, 838–842 (2000). expression programs controlling breast cancer pathogenesis. Genome Res. 66. Hintermair, C. et al. Treonine-4 of mammalian RNA polymerase II CTD 28, 159–170 (2018). is targeted by Polo-like kinase 3 and required for transcriptional elongation. 36. Hah, N., Murakami, S., Nagari, A., Danko, C. G. & Kraus, W. L. Enhancer EMBO J. 31, 2784–2797 (2012). transcripts mark active estrogen receptor binding sites. Genome Res. 23, 67. Hsin, J. P. & Manley, J. L. Te RNA polymerase II CTD coordinates 1210–1223 (2013). transcription and RNA processing. Genes Dev. 26, 2119–2137 (2012). 37. Core, L. J. et al. Analysis of nascent RNA identifes a unifed architecture of 68. Mayer, A. et al. CTD tyrosine phosphorylation impairs termination factor initiation regions at mammalian promoters and enhancers. Nat. Genet. 46, recruitment to RNA polymerase II. Science 336, 1723–1725 (2012). 1311–1320 (2014). 69. Corden, J. L. Tails of RNA polymerase II. Trends Biochem. Sci. 15, 38. Hah, N. et al. Infammation-sensitive super enhancers form domains of 383–387 (1990). coordinately regulated enhancer RNAs. Proc. Natl Acad. Sci. USA 112, 70. West, M. L. & Corden, J. L. Construction and analysis of yeast RNA E297–E302 (2015). polymerase II CTD deletion and substitution . 140, 39. Hah, N. et al. A rapid, extensive, and transient transcriptional response to 1223–1233 (1995). estrogen signaling in breast cancer cells. Cell 145, 622–634 (2011). 71. Eick, D. & Geyer, M. Te RNA polymerase II carboxy-terminal domain 40. Kwak, H., Fuda, N. J., Core, L. J. & Lis, J. T. Precise maps of RNA (CTD) code. Chem. Rev. 113, 8456–8490 (2013). polymerase reveal how promoters direct initiation and pausing. Science 339, 72. Bonn, S. et al. Tissue-specifc analysis of chromatin state identifes temporal 950–953 (2013). signatures of enhancer activity during embryonic development. Nat. Genet. 41. Mahat, D. B. et al. Base-pair-resolution genome-wide mapping of active 44, 148–156 (2012). RNA polymerases using precision nuclear run-on (PRO-seq). Nat. Protoc. 73. Descostes, N. et al. Tyrosine phosphorylation of RNA polymerase II CTD is 11, 1455–1476 (2016). associated with antisense promoter transcription and active enhancers in 42. Carninci, P. et al. Genome-wide analysis of mammalian promoter mammalian cells. Elife 3, e02105 (2014). architecture and evolution. Nat. Genet. 38, 626–635 (2006). 74. Hsin, J. P., Li, W., Hoque, M., Tian, B. & Manley, J. L. RNAP II CTD tyrosine 43. Shiraki, T. et al. Cap analysis gene expression for high-throughput analysis 1 performs diverse functions in vertebrate cells. Elife 3, e02112 (2014). of transcriptional starting point and identifcation of promoter usage. 75. Bentley, D. L. Coupling mRNA processing with transcription in time and Proc. Natl Acad. Sci. USA 100, 15776–15781 (2003). space. Nat. Rev. Genet. 15, 163–175 (2014). 44. Arner, E. et al. Transcribed enhancers lead waves of coordinated 76. Shah, N. et al. Tyrosine-1 of RNA polymerase II CTD controls global transcription in transitioning mammalian cells. Science 347, termination of gene transcription in mammals. Mol. Cell 69, 1010–1014 (2015). 48–61.e6 (2018).

Nature Structural & Molecular Biology | www.nature.com/nsmb Review Article NAtuRe StRuctuRAL & MoLecuLAR BioLogy

77. Nojima, T. et al. Deregulated expression of mammalian lncRNA through 105. Hsieh, C.-L. et al. Enhancer RNAs participate in androgen receptor-driven loss of SPT6 induces R-loop formation, replication stress, and cellular looping that selectively enhances gene activation. Proc. Natl Acad. Sci. USA senescence. Mol. Cell 72, 970–984.e7 (2018). 111, 7319–7324 (2014). 78. Wang, A. H. et al. Te elongation factor Spt6 maintains ESC pluripotency 106. Mousavi, K. et al. eRNAs promote transcription by establishing chromatin by controlling super-enhancers and counteracting Polycomb proteins. accessibility at defned genomic loci. Mol. Cell 51, 606–617 (2013). Mol. Cell 68, 398–413.e6 (2017). 107. Sigova, A. A. et al. Transcription factor trapping by RNA in gene regulatory 79. Lin, C., Garruss, A. S., Luo, Z., Guo, F. & Shilatifard, A. Te RNA Pol II elements. Science 350, 978–981 (2015). elongation factor Ell3 marks enhancers in ES cells and primes future gene 108. Zhao, Y. et al. Activation of P-TEFb by androgen receptor-regulated activation. Cell 152, 144–156 (2013). enhancer RNAs in castration-resistant prostate cancer. Cell Rep. 15, 80. Gil, N. & Ulitsky, I. Production of spliced long noncoding RNAs specifes 599–610 (2016). regions with increased enhancer activity. Cell Syst. 7, 537–547.e3 (2018). 109. Tan, S. H. et al. Te enhancer RNA ARIEL activates the oncogenic 81. Vučićević, D., Corradin, O., Ntini, E., Scacheri, P. C. & Ørom, U. A. Long transcriptional program in T-cell acute lymphoblastic leukemia. Blood 134, ncRNA expression associates with tissue-specifc enhancers. Cell Cycle 14, 239–251 (2019). 253–260 (2015). 110. Aguilera, A. & García-Muse, T. R loops: from transcription byproducts to 82. Core, L. J. et al. Defning the status of RNA polymerase at promoters. threats to genome stability. Mol. Cell 46, 115–124 (2012). Cell Rep. 2, 1025–1035 (2012). 111. Meng, F. L. et al. Convergent transcription at intragenic super-enhancers 83. Kaikkonen, M. U. et al. Remodeling of the enhancer landscape during targets AID-initiated genomic instability. Cell 159, 1538–1548 (2014). macrophage activation is coupled to enhancer transcription. Mol. Cell 51, 112. Aguilo, F. et al. Deposition of 5-methylcytosine on enhancer RNAs enables 310–325 (2013). the coactivator function of PGC-1α. Cell Rep. 14, 479–492 (2016). 84. Andersson, R. et al. Nuclear stability and transcriptional directionality 113. Maris, C., Domínguez, C. & Allain, F. H. Te RNA recognition motif, a separate functionally distinct RNA species. Nat. Commun. 5, 5336 (2014). plastic RNA-binding platform to regulate post-transcriptional gene 85. Pefanis, E. et al. RNA exosome-regulated long non-coding RNA expression. FEBS J. 272, 2118–2131 (2005). transcription controls super-enhancer activity. Cell 161, 774–789 (2015). 114. Grishin, N. V. KH domain: one motif, two folds. Nucleic Acids Res. 29, 86. Lubas, M. et al. Te human nuclear exosome targeting complex is loaded 638–643 (2001). onto newly synthesized RNA to direct early ribonucleolysis. Cell Rep. 10, 115. Chang, K. Y. & Ramos, A. Te double-stranded RNA-binding motif, a 178–192 (2015). versatile macromolecular docking platform. FEBS J. 272, 2109–2117 (2005). 87. Austenaa, L. M. et al. Transcription of mammalian cis-regulatory elements 116. Nair, S. J. et al. Phase separation of ligand-activated enhancers licenses is restrained by actively enforced early termination. Mol. Cell 60, cooperative chromosomal enhancer assembly. Nat. Struct. Mol. Biol. 26, 460–474 (2015). 193–203 (2019). 88. Ulitsky, I. & Bartel, D. P. lincRNAs: genomics, evolution, and mechanisms. 117. Banani, S. F., Lee, H. O., Hyman, A. A. & Rosen, M. K. Biomolecular Cell 154, 26–46 (2013). condensates: organizers of cellular biochemistry. Nat. Rev. Mol. Cell Biol. 18, 89. Guttman, M. & Rinn, J. L. Modular regulatory principles of large 285–298 (2017). non-coding RNAs. Nature 482, 339–346 (2012). 118. Ruthenburg, A. J., Li, H., Patel, D. J. & Allis, C. D. Multivalent engagement 90. Cabili, M. N. et al. Integrative annotation of human large intergenic of chromatin modifcations by linked binding modules. Nat. Rev. Mol. Cell noncoding RNAs reveals global properties and specifc subclasses. Biol. 8, 983–994 (2007). Genes Dev. 25, 1915–1927 (2011). 119. Hnisz, D., Shrinivas, K., Young, R. A., Chakraborty, A. K. & Sharp, P. A. A 91. Almada, A. E., Wu, X., Kriz, A. J., Burge, C. B. & Sharp, P. A. Promoter phase separation model for transcriptional control. Cell 169, 13–23 (2017). directionality is controlled by U1 snRNP and polyadenylation signals. 120. Polymenidou, M. Te RNA face of phase separation. Science 360, Nature 499, 360–363 (2013). 859–860 (2018). 92. Paralkar, V. R. et al. Unlinking an lncRNA from its associated cis element. 121. Boija, A. et al. Transcription factors activate genes through the Mol. Cell 62, 104–110 (2016). phase-separation capacity of their activation domains. Cell 175, 93. Grof, A. F. et al. In vivo characterization of Linc-p21 reveals functional 1842–1855.e16 (2018). cis-regulatory DNA elements. Cell Rep. 16, 2178–2186 (2016). 122. Sabari, B. R. et al. Coactivator condensation at super-enhancers links phase 94. Kopp, F. & Mendell, J. T. Functional classifcation and experimental separation and gene control. Science 361, eaar3958 (2018). dissection of long noncoding RNAs. Cell 172, 393–407 (2018). 123. Boeynaems, S. et al. Protein phase separation: a new phase in cell biology. 95. Alvarez-Dominguez, J. R., Knoll, M., Gromatzky, A. A. & Lodish, H. F. Te Trends Cell Biol. 28, 420–435 (2018). super-enhancer-derived alncRNA-EC7/Bloodlinc potentiates red blood cell development in trans. Cell Rep. 19, 2503–2514 (2017). Acknowledgements 96. Espinosa, J. M. On the origin of lncRNAs: missing link found. We thank Jim Kadonaga for reading the manuscript and providing helpful suggestions. Trends Genet. 33, 660–662 (2017). Research in the Lauberth laboratory is supported by a grant from the NIH/National 97. Cinghu, S. et al. Intragenic enhancers attenuate host gene expression. Institute of General Medical Sciences (R35 GM128900) to S.L., and research in Mol. Cell 68, 104–117.e6 (2017). the Sartorelli laboratory is supported by the National Institute of Arthritis and 98. Shechner, D. M., Hacisuleyman, E., Younger, S. T. & Rinn, J. L. Musculoskeletal and Skin Diseases Intramural Research Program of the NIH (grants Multiplexable, locus-specifc targeting of long RNAs with CRISPR-Display. AR041126 and AR041164). Nat. Methods 12, 664–670 (2015). 99. Lee, J. S. & Mendell, J. T. Antisense-mediated transcript knockdown triggers premature transcription termination. Mol. Cell 77, 1044–1054.e3 (2020). Competing interests 100. Lai, F., Damle, S. S., Ling, K. K. & Rigo, F. Directed RNase H cleavage of The authors declare no competing interests. nascent transcripts causes transcription termination. Mol. Cell 77, 1032–1043.e4 (2020). 101. Yang, Y. et al. Enhancer RNA-driven looping enhances the transcription of Additional information the long noncoding RNA DHRS4-AS1, a controller of the DHRS4 gene Correspondence should be addressed to S.M.L. cluster. Sci. Rep. 6, 20961 (2016). Peer review information Beth Moorefield was the primary editor on this article and 102. Chu, C., Qu, K., Zhong, F. L., Artandi, S. E. & Chang, H. Y. Genomic maps managed its editorial process and peer review in collaboration with the rest of the of long noncoding RNA occupancy reveal principles of RNA-chromatin editorial team. interactions. Mol. Cell 44, 667–678 (2011). Reprints and permissions information is available at www.nature.com/reprints. 103. Femino, A. M., Fay, F. S., Fogarty, K. & Singer, R. H. Visualization of single RNA transcripts in situ. Science 280, 585–590 (1998). Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in 104. Bose, D. A. et al. RNA binding to CBP stimulates histone acetylation and published maps and institutional affiliations. transcription. Cell 168, 135–149.e22 (2017). © Springer Nature America, Inc. 2020

Nature Structural & Molecular Biology | www.nature.com/nsmb